WhichAI

WhichAI field guides

100 ways to turn AI into operating capacity.

Most AI content stops at a tool list. These guides map the underlying work, the system boundary, the human review, the controls, and the scorecard. The target is not more output from a chatbot. The target is making the team you already have capable of handling more work without hiding a broken process inside another hire.

The WhichAI test

If the human still spends the day collecting, copying, chasing, formatting, and routing, the implementation is not finished.

A complete system prepares the work, preserves evidence, routes exceptions, and gives the human a clean decision surface.

Editorial method: Editorial synthesis of workflow design patterns and implementation constraints. Public control references provide context, not proof of a deployment or legal advice. Where a versioned evidence pack appears, its evidence class, method, and limitations govern what the artifact can support. Read the full method. By WhichAI; published by WhichAI.

The task is already clear

Get the full workflow recommendation from self-serve WhichAI

Describe the exact job. Get the tools, prompts, configuration steps, workflow sequence, pricing context, and explicit economic assumptions in one workflow.

Plan my workflow

The workflow is becoming headcount

Bring the operating problem to WhichAI Solutions

Map the work beneath the role, find the leak, and decide whether the answer is a hire, a tool, or a narrower operating system.

See company solutions

Work, not hype

Every guide starts with the queue, handoff, exception, or review cycle consuming capacity.

The whole system

Inputs, source systems, routing, human gates, evidence, and failure handling stay in the same design.

A better human job

The system removes preparation and coordination work so the person keeps judgment, approval, and exception ownership.

Controls before scale

Each design names the records, access boundaries, review points, and measures required before more volume is trusted to it.

Evidence-backed implementation guides

Inspect the artifact, method, and limitation behind the recommendation

These guides publish templates, synthetic fixtures, or observed product output. None is presented as a customer result.

Hire or automate an operations role

Editable decision matrix and a worked synthetic scenario.

Inspect the evidence

AI saves time only when work disappears

Public Work Removal Ledger and synthetic cases.

Inspect the evidence

What WhichAI returns for a real task

Archived deterministic planner capture, browser screenshot, vendor check, and annotations.

Inspect the evidence

Map a workflow before proposing AI

Blank current-state process map, labeled synthetic example, and mapping protocol.

Inspect the evidence

Research and evidence

Inspect the artifact, method, sources, and limitations.

The research library separates templates, synthetic fixtures, observed outputs, and test protocols so the evidence is not mistaken for a customer result or guaranteed outcome.

Open the research library

The library

Start with the work that is eating the team.

Ten clusters. Ten field guides in each. Every page answers a specific implementation or headcount decision.

01

Real AI Implementation Creates Capacity

Field guides for distinguishing AI activity from systems that remove recurring preparation, coordination, and exception work from the human queue.

Evaluate whether an AI initiative changes throughput, review load, handoffs, and operating capacity instead of merely adding another tool.

01AI Does Not Save Time Until Work DisappearsA work-removal scorecard for separating faster individual tasks from recurring work that has actually left the operating queue.02AI Adoption vs AI Implementation: Buying Tools Is Not the Same as Removing WorkA maturity model for tracing the distance between tool access, repeated usage, a controlled workflow, and dependable operating capacity.03Headcount Is Not Throughput: Why More People Do Not Fix a Broken WorkflowA queue and handoff analysis for determining whether backlog comes from labor demand, work design, system delay, rework, or exception ownership.04Can One Operator Absorb the Work Behind Four Open Requisitions?An illustrative task-volume model for testing how preparation systems could change workload without promising role replacement or staffing outcomes.05Why Adding an AI Tool Can Create More Work Than It RemovesA before-and-after handoff ledger for exposing prompt preparation, output repair, copying, approvals, and support work introduced by an AI tool.06Automate the Queue, Not the Job TitleA task, judgment, and exception decomposition for redesigning a recurring queue without claiming that a system replaces a person or profession.07Human Review Is Not a Failure of AutomationA risk-tiered review design for directing human attention to consequential, ambiguous, low-confidence, and exception cases without recreating the original workload.08The Exception Path Is the Real WorkflowAn exception operating model for detecting missing, conflicting, unusual, or failed cases and routing them with evidence to a named owner.09From Tool Stack to Operating Capacity: The Missing Implementation LayerA stack-to-owner-to-metric map for turning selected products into one controlled workflow with source evidence, review, and recovery.10Five Tests That Separate Real AI Implementation From an AI DemoA five-test readiness rubric for evidence, variation, human judgment, failure recovery, and measurable downstream acceptance.
02

Hire, Tool, or Rebuild

Decision frameworks for mapping the work beneath a role before funding another person, another platform, or a broad rebuild.

Decide whether a specific operations constraint needs human capacity, a bounded workflow system, a process redesign, or no project.

01Should I Hire or Automate This Operations Role?A hire-or-fix decision tree that separates recurring preparation from judgment, service, exception ownership, and genuine labor demand.02Hire vs Tool vs Workflow Rebuild: A Decision Framework for COOsA weighted COO decision matrix for matching labor, software, process redesign, or no project to the measured source of operating drag.03Audit the Workflow Before You Approve the Job RequisitionA pre-requisition audit for tracing cases, wait states, systems, rework, exceptions, and judgment before funding another recurring operations role.04Why You Keep Hiring for the Same RoleA repeated-role root-cause analysis for separating growth, turnover, service demand, and mechanical workflow drag before opening the role again.05Operations Headcount Planning When AI Can Remove Mechanical WorkAn AI-aware capacity plan that models task mix, volume, review, exceptions, coverage, and scenario uncertainty without presenting projected savings as results.06Backlog Math: When a Staffing Problem Is Really a Flow ProblemA queue diagnosis that separates arrival rate, service rate, wait states, rework, preparation, and review capacity before another staffing request.07Build vs Buy vs Hire for Business Process AutomationA common-scope comparison of internal build, packaged software, services, labor, and no project using total operating cost and residual work.08How to Run a 90-Day Workflow Pilot Before Adding HeadcountA reversible ninety-day pilot charter for testing one preparation slice, one owner, one queue, and one staffing-relevant scorecard.09What Work Must Stay Human in an AI-Enabled Operation?A consequence and accountability map for keeping judgment, approval, communication, recourse, and exception ownership with the right people.10When AI Automation Is the Wrong AnswerA no-project test for low-volume, unstable, inaccessible, high-consequence, or unowned workflows where process repair or human capacity fits better.
03

Self-Serve AI Workflow Blueprints

Task-specific guides for designing a stack, setup order, evidence path, review gates, fallbacks, costs, and handoffs before implementation.

Turn a known business task into a concrete, model-agnostic workflow blueprint or identify when the work needs a scoped implementation.

01Which AI Tools Should I Use for This Exact Business Task?A task-first selection method for choosing current tool categories, handoffs, evidence, review, fallback, and cost assumptions for one job.02How to Build an AI Stack Without Opening 40 Vendor TabsA requirement-led shortlist that reduces vendor research to the few components that fit one task, one data boundary, and one review design.03Generic Chatbot vs Operating Workflow: A Reproducible Comparison ProtocolA frozen task, blank rubric, and reproducible protocol for testing what must be added to chatbot advice before recurring work can run with evidence, ownership, and recovery.04AI Workflow Blueprint: What Must Be Included Before You BuildA build-readiness checklist for the task contract, source data, component roles, handoffs, human review, exceptions, controls, costs, and pilot evidence.05From One Sentence to a Full AI Tool Stack: How Workflow Design WorksAn annotated task decomposition showing how a short request becomes a specific input, output, component, evidence, review, exception, and measurement plan.06How to Combine AI Models, Specialized Tools, and Automation SoftwareA model-agnostic component design for assigning retrieval, generation, transformation, routing, storage, evidence, and review to the right layer.07The Right Account Setup Order for an AI WorkflowA dependency graph for sequencing approved vendor configuration, roles, data boundaries, test records, integrations, review, billing, and rollback checks.08How to Evaluate an AI Workflow Generator Before You PayA reproducible buyer test for task specificity, current sources, setup detail, evidence, review, failure handling, cost assumptions, and usable output.09Fallbacks, Handoffs, and Review Gates: The Parts Most AI Plans Leave OutA workflow hardening guide for defining record transfer, human authority, failure ownership, manual continuity, reconciliation, and rollback.10What a Former WhichAI Intake Prototype Returned for One Frozen TaskAn inspectable historical capture of a former WhichAI intake prototype, including the exact input, deterministic output, screenshot, method, and limitations.
04

Healthcare and HIPAA-Aware AI Workflows

Compliance-aware healthcare workflow designs that require BAA verification, PHI data-flow mapping, organizational risk review, safeguards, and accountable human judgment.

Research healthcare workflow preparation while preserving PHI controls, contractual review, risk ownership, and clinical or administrative approval.

01Can You Use AI With PHI? Start With the Data FlowA compliance-aware PHI data-flow assessment covering purpose, minimum fields, BAAs, vendors, subprocessors, storage, logs, safeguards, and human approval.02What Makes an AI Workflow Compliance-Aware Under HIPAA?A requirements matrix for connecting the actual PHI workflow to contracts, safeguards, organizational risk review, human authority, and lifecycle evidence.03AI Vendor BAA Checklist for Healthcare OperationsA dated vendor-chain checklist for matching BAAs, permitted PHI use, subprocessors, safeguards, service terms, return, deletion, and breach responsibilities to the workflow.04Map Every Place PHI Touches Before Connecting an AI ToolA field-level system-boundary map for finding PHI copies in prompts, APIs, logs, caches, storage, support, exports, backups, and downstream applications.05Healthcare Intake Automation Without Automating Clinical JudgmentA compliance-aware intake preparation workflow that collects, validates, and routes PHI while clinicians retain triage, diagnosis, treatment, and escalation judgment.06Revenue Cycle Automation: Remove Re-Keying Before Adding Billing StaffA compliance-aware RCM preparation design for reconciling source facts, payer requirements, work queues, and denial evidence before authorized billing review.07Prior Authorization Workflow Automation: What AI Can Prepare and Humans Must ApproveA compliance-aware prior authorization packet workflow for collecting cited evidence and payer requirements while authorized people retain clinical and submission decisions.08Clinical Documentation AI: Account Setup, Data Controls, and Human ReviewA compliance-aware documentation draft workflow for approved access, minimum PHI, source preservation, clinician correction, record approval, and lifecycle control.09Healthcare Document Automation: OCR Is Not the Whole WorkflowA compliance-aware document lifecycle adding source custody, classification, validation, exception routing, human approval, acknowledgement, and deletion to OCR.10Healthcare AI Blueprint or Implementation Partner: Which Path Fits the Workflow?A fit matrix for choosing self-serve planning, a scoped diagnostic, a controlled pilot, or no project based on PHI, systems, judgment, ownership, and risk.
05

Property Management Capacity

Workflow designs that move documents, assemble packages, route follow-up, and preserve accountable decisions in property operations.

Property managers evaluating practical automation without delegating tenant, legal, or safety judgment to software.

01Property Management Automation: Start With Lease Packages, Not Tenant DecisionsMap document intake, completeness checks, lease assembly, and approval before considering automation near tenant eligibility decisions.02Applicant Document Intake Workflow for Property ManagersBuild a controlled intake queue that identifies missing applicant files, preserves originals, and routes exceptions for staff review.03Automate Lease Package Assembly Without Losing Human ApprovalAssemble lease fields, disclosures, and attachments from verified sources while keeping material terms under named human approval.04Maintenance Request Triage: Route Faster Without Letting AI Make Safety DecisionsStructure maintenance intake and routing while named people retain emergency, habitability, access, and vendor safety decisions.05Stop Chasing Vendor COIs by EmailTurn certificate requests, receipt tracking, field review, expiration reminders, and exceptions into one property vendor queue.06Property Management Renewal Prep as a Repeatable WorkflowPrepare renewal context, dates, property policy, communications, and approvals without delegating the renewal decision to software.07Automate Owner Reporting Without Rebuilding the Report Every MonthCreate owner reports from governed property sources, explain exceptions, and retain approval without reconstructing the package monthly.08Inspection Follow-Up Automation for Property Management TeamsTurn inspection findings into owned actions, evidence, due dates, resident coordination, and escalations instead of scattered notes.09Delinquency Communication Workflows: Automate Preparation, Preserve Legal ReviewPrepare accurate balances, timelines, and message drafts while authorized staff retain legal review, tenant treatment, and delivery decisions.10Do You Need Another Leasing Coordinator or a Better Workflow?Decompose the leasing queue before hiring so the company can separate mechanical preparation from tenant-facing judgment and service.
06

Logistics and 3PL Capacity

Evidence-led freight workflows that turn scattered documents, events, and exceptions into billing-ready and service-ready work.

Logistics leaders looking to remove document chase, status chase, and billing preparation work from daily queues.

01Proof-of-Delivery Automation: Turn Delivered Loads Into Billing-Ready FilesConnect delivered-load events, POD collection, document validation, exceptions, and invoice handoff without hiding missing evidence.02How to Automate Missing POD Follow-Up Without Losing the Exception QueueAutomate factual POD requests and reminders while preserving carrier responses, disputed cases, aging, and human escalation ownership.03Rate Confirmation Data Entry Is a Workflow ProblemExtract rate confirmation terms into a reviewable TMS handoff with source coordinates, validation rules, and a human exception queue.04Accessorial Charge Automation: Capture the Evidence Before BillingCreate an evidence-first accessorial workflow for event capture, document collection, rule review, approval, and billing handoff.05Detention Tracking Without Another SpreadsheetTrack appointment, arrival, release, threshold, evidence, and escalation events in one load-linked workflow instead of a detached sheet.063PL Invoice Discrepancy Triage With Human ApprovalClassify invoice differences, assemble contract and shipment evidence, and route proposed resolutions to accountable reviewers.07Automated Shipment Status Updates That Do Not Invent an ETAGenerate customer updates only from verified shipment events, with explicit uncertainty, source timestamps, and human exception review.08Carrier Onboarding Automation: Documents, Checks, and Exception OwnershipCoordinate carrier intake, document collection, source checks, approvals, and renewals without treating an automated check as final authority.09Appointment Scheduling Automation for High-Volume Freight OperationsCoordinate facility requests, load constraints, confirmed slots, changes, and fallback queues without inventing availability or confirmation.10Hire Another Billing Coordinator or Fix POD-to-Cash?Measure POD collection, evidence review, exception chase, invoice preparation, and dispute work before adding billing headcount.
07

Commercial Insurance Operations

Human-reviewed insurance workflows for intake, document preparation, renewal, certificates, claims support, and quote comparison.

Insurance operators seeking automation around licensed judgment, consumer impact, and source-level evidence requirements.

01Commercial Insurance Workflow Automation: Where to StartPrioritize intake, document, follow-up, and preparation workflows while licensed and accountable people retain insurance judgment.02First Notice of Loss Intake: Automate Collection, Not Coverage DecisionsStructure FNOL facts, documents, source records, and escalation while adjusters retain coverage, liability, reserve, and claim decisions.03Insurance Submission Triage Without Autonomous UnderwritingPrepare commercial submissions for appetite and completeness review while underwriters retain risk selection, pricing, and terms.04Extract Policy Data From PDFs Without Trusting Every FieldUse field-level confidence, page coordinates, validation, and review before policy facts enter downstream insurance workflows.05Renewal Preparation Automation for Commercial Insurance TeamsAssemble current exposure, policy, loss, service, and market context into a reviewed renewal packet with source traceability.06COI Request and Follow-Up Automation for BrokersCoordinate certificate requests, policy source checks, issuance preparation, delivery, and exceptions without inventing coverage facts.07Claims Document Chase Work Is Consuming Adjuster CapacitySeparate document request, receipt, matching, completeness, and escalation work from the claim judgment adjusters must retain.08Quote Comparison Workflows With Source-Level TraceabilityCompare commercial insurance quote fields with document lineage, normalization rules, conflict review, and qualified client advice.09Broker Operations Automation: Connect Intake, Renewal, COI, and Follow-UpConnect broker service queues through shared account identity, source records, review gates, exceptions, and accountable handoffs.10Do You Need Another CSR or a Better Commercial Insurance Workflow?Decompose account service workload into preparation, follow-up, exceptions, client service, and licensed judgment before hiring.
09

Title and Mortgage Operations

Evidence-preserving workflows for title requirements, mortgage documents, closing packages, recording follow-up, and pipeline coordination.

Design title and mortgage operations workflows that reduce document chase while preserving legal, credit, and quality review boundaries.

01Title Processing Automation: Start With Stipulation TrackingA stipulation-tracking workflow that ties every title requirement to its source, owner, evidence, exception state, and closing readiness decision.02Title Exception Triage With Human Legal ReviewA source-linked title exception workflow that organizes facts, evidence, priority, and ownership while authorized reviewers retain legal disposition.03Closing Package Assembly Without Manual Document ChaseA closing package workflow that turns a versioned checklist, controlled document intake, validation, and final approval into one traceable production process.04Mortgage Document Intake: Extract, Validate, and EscalateA mortgage document intake workflow for file identity, structured extraction, cross-source validation, exception queues, and reviewer handoff.05Automate Borrower Document Follow-Up Without Making Credit DecisionsA borrower follow-up workflow driven by explicit missing-document states, approved communication, pause rules, and human lending review.06Pre-Closing Quality Control as an AI-Assisted WorkflowA pre-closing quality-control design that reconciles approved transaction data, documents, exceptions, and signatures before an authorized readiness decision.07Post-Close Document Review Without a Growing Cleanup TeamA post-close workflow for package inventory, indexing, defect preparation, correction ownership, and archive evidence without hiding reviewer load.08Lien and Recording Follow-Up Automation With Source TraceabilityA follow-up workflow that ties each lien or recording task to the authoritative instrument, jurisdiction source, owner, status event, and closure evidence.09Mortgage Pipeline Status Reporting Without Manual ReconciliationA pipeline reporting workflow that maps source-system events into defined milestones, exposes conflicts, and preserves the reason behind each status.10Hire Another Processor or Fix the Mortgage Workflow?A hire-versus-fix analysis for separating processor judgment from document intake, status reconciliation, follow-up, and repeated file preparation.
10

Measurement Governance and Proof

Practical methods for baselines, human review, audit trails, vendor diligence, pilots, reliability decisions, and operating-capacity evidence.

Measure and govern AI workflows with local evidence instead of relying on demos, generic benchmarks, or unsupported outcome claims.

01How to Measure Time Saved by an AI WorkflowA practical measurement design for comparing human work before and after an AI workflow without ignoring review, corrections, exceptions, or displaced labor.02Build the Baseline Before You Automate AnythingA baseline-building workflow for process boundaries, volume, time, quality, exceptions, cost, and evidence before an automation pilot changes the operation.03Exception Rate Is the AI Metric Most Teams IgnoreA workflow measurement design for defining, capturing, aging, resolving, and learning from the cases an AI system cannot handle cleanly.04Designing Human Review for AI WorkflowsA human-review design for risk tiers, decision-ready evidence, explicit dispositions, sampling, escalation, and feedback that improves the workflow.05What an AI Workflow Audit Trail Must CaptureA practical audit-trail schema for identity, access, inputs, sources, configuration, tool actions, human decisions, changes, and retention.06AI Vendor Due Diligence: Pricing, Data, Retention, Contracts, and Failure ModesA vendor evaluation workflow that maps real usage economics, data paths, retention, contract terms, controls, dependencies, failures, and exit requirements.07How to Pilot an AI Workflow Without Turning the Company Into a Migration ProjectA bounded pilot design that uses representative work, shadow operation, explicit interfaces, human review, rollback, and a written scale decision.08When Is an AI Workflow Reliable Enough to Scale?A scale-readiness gate based on an explicit acceptance envelope, representative reliability evidence, severe-failure tests, recovery, operating load, and drift controls.09The Failure Modes Every AI Workflow Blueprint Should NameA practical failure register for source, model, integration, access, human-review, queue, downstream-action, and change failures, with detection and fallback attached.10How to Prove an AI Implementation Created Real Operating CapacityAn executive proof method that separates observed time, throughput, quality, review, and queue changes from modeled capacity and headcount implications.